Maybe you are a specialist in two-phase problems and know which model, two-fluid model and mixture model, is better for bubbly flows. It seems that many people use two-fluid models, especially for engineering problems. But this kind of models has drawbacks, e.g., it needs two sets of governing equations, one for gas and the other for water, and thus is too complicated. On the other hand, mixture models seem to be less complicated, but I have not found some papers about their turbulence models. If you know something about this, please let me know. Thanks a lot.

It depends on what you are looking for with the model. Two-fluid model can keep track of the movement and heat/mass transfer within each phase and also be able to calculate the velocity and heat/mass movement on the phase interface, the later of which is sometimes very important, or the focus of the study. The mixture model is simpler but does not give you the interface situation. Often, a seperate local formulations can be used (coupled with or based on the result from the mixture model) to get some limited prediction of the local parameters. Again, it all depends what you are looking for.

Yes, I agree with; a mixture model cannot describe interface phenomena in details. Actually the difficulty for a good understanding of a two-phase flow araises from the phenomena. However, currently we are interested only in concentration of gas in water, and the mixture model seems to be nice. What I feel to be not sound for this choice is that no proper turbulence model is available.

Sir, do you know some work about the turbulence model? By the way, can you tell me some good references about the separate local formulations you mentioned? Thanks.

I am not sure what you mean by mixture model. If it means the homogeneous mixture model (i.e. no slip, phases well mixed, etc.) I think that it does not make a lot of sense to look for a special two-phase flow turbulence model. You can just use a single-phase model with apparant density, average velocity, etc. equivalent to that of two-phase mixture.

Are you interested in average concentration or may be you want to know concentration profile. In the later case I would suggest using a two-fluid model.

You may want to check the following authors: Serizawa, Bataille, Lopez de Bertodano in any scientifc database (e.g. Compendex). If you screen through some issues of the International Jounal of Multiphase Flow you may find out interesting articles.

Andrzej Matuszkiewicz

Dear Andrzej,

What I mean by a mixture model is often refereed as to a pseudo-fluid model, which considers a bubbly flow as one pseudo-fluid by introducing a pseudo-density that varies between gas and water density extremes. One example of such model can be found in J. Fluid Mech., v240, p59, by Kubota etc.

For such a model I cannot see the reason to simply apply one-phase turbulence model. At least, in the well known k-epsilon model, the constants are determined by some, althogth simple, experimental data. However, can we also find some experiment results for a mixture model?

As I mentioned in the last message, a main problem to apply a pseudo-fluid method for practical problems is to develop a proper turbulence model.

Any person has idea about such development?

Thanks.

[arif940]

Hi, as you said, how can I do two liquid flow simulation and find concentration profile for mixture. Both liquid have inlet velocity. Thanks